The Future of Virtual Personal Assistants

Mike Elgan has a good write up of where things will hopefully go in the world of Virtual Personal Assistants (VPAs). Essentially, he’s recommending two things.

Bring Your Own Assistant (BYOA)

First, employees should be free to choose their own VPA for use at work, rather than forcing some sort of corporate standard. We don’t, in other words, want our employer telling us “OK, everyone, it’s decided: we’re all going to use Siri.” As Mike puts it:

In other words, virtual assistants used in business will not be developed or mandated by the company, but chosen by the employee and carried from one job to the next. This vision is almost certainly accurate, although I would also imagine that companies will deploy and even build their own assistants to add to those chosen and used by employees.

This totally makes sense. Companies that don’t move in this direction will eventually be at a severe disadvantage when it comes to attracting the best talent. Why is that? Because, over time, each of us will form very tight, essentially symbiotic, relationships with our VPA. The VPA will get to know our preferences on an intimate level through hours and hours of observational training in serving us. Our relationship with our VPA will be a core aspect of what actually shows up at work and an important part of what we contribute to organizations. Stripping that away, we actually subtract value from our places of work.

Furthermore, in an economy where the average stay in any one organization is getting shorter and shorter, and large portions of the workforce work exclusively in a gig economy, the thought of any one organization dictating something as personal as our choice of VPA will eventually seem laughable.

Hide the Complexity

Mike’s second point is that the app-like framework used today for plugging services into a VPA is pretty broken:

Virtual assistants should work initially like apps, then later not at all like apps. At first individual users should find, and in some cases pay for or subscribe to, specific virtual assistant services such as a meeting scheduler or a flight-booking add-ons.

Once installed, however, the user of those services should be the virtual assistant, not the human.

This is right on. Forcing users to remember what skills are available and then asking them to choose a specific skill provider is merely a leftover behavior from today’s visual user interfaces. As more skills and services plug into the VPA market, we would end up like warlocks and witches, memorizing incantations to accomplish this or that task. That seems broken.

The obvious answer is that this is not the way things will evolve. As we move into an increasingly conversational user interface, the vast majority of that complexity will simply be hidden away. A big part of what our VPA will be learning from us is our preferences in choosing third-party skills and services. Are we value-oriented, always looking for the least expensive solution? Do we prioritize comfort, speed, or social acceptance? How does one particular skill map to the other skills we’ve come to trust over time?

There is lots of nuance here and it’s all part of the personalization that will be an integral part of a Virtual Personal Assistance getting to know us. It’s part of the training process. We will set the objectives and the parameters, and then automate the execution. And part of that execution will eventually include selecting the skill provider.

No more need for incantations.



Leave a Reply